UO lab advances research on memory formation and recall

Illustration of human head with file drawer extending from brain

Difficulty remembering a password or where a car is parked outside a store doesn’t necessarily suggest memory problems. It’s likely that interference is clouding the brain, suggests University of Oregon neuroscientist Brice Kuhl.

How do people encode new information and accurately recall it later? That’s the focus of the Kuhl lab. His research is basic, or fundamental, science. Before treating disease, scientists first need to understand normal functioning, said Kuhl, an associate professor in the Department of Psychology and a member of the Institute of Neuroscience.

“We don’t study disease states, but the brain regions we are looking at are often compromised in disease states such as Alzheimer’s disease,” he said. “We study memory interference, when memories are similar and overlap with others, and we have trouble keeping them separate. It’s a recipe for confusion.”

In two recent studies, Kuhl’s lab has gleaned new insights from experiments with human subjects doing simple tasks and using magnetic resonance imaging. MRI scans help reveal how memories are represented in activity patterns in the brain’s ventral parietal cortex, which is vital to forming and recalling memories.

In Current Biology, a study led by Kuhl’s former postdoctoral researcher Nicole M. Long found that activity patterns related to past memories tend to take precedence over incoming new but similar information. That activity, it was noted, may represent an in-the-moment occurrence not necessarily tied to stronger and deliberate reactivation of memories, which the project did not explore.

The study reinforces previous findings in neuroscience that found neurons in the parietal cortex of rodents doing a new task continued to reflect activity related to previous similar experiences.

“This study tells us that there are some parts of the brain — maybe not specific to memory, maybe just to internal thoughts — that show stronger representations of the past than information from the present,” Kuhl said. “This suggests some kind of memory specificity that we find interesting.”

In the project, Long, now an assistant professor of psychology at the University of Virginia, initially showed 33 participants photos of 24 different objects representing differing categories while they were being scanned by MRI. The subjects subsequently viewed 24 new but different objects from the same groupings. Subjects were nudged to either think of the new object or the similar previously viewed object. As in the rodents, the previous objects dominated the new ones in the parietal cortex.

In a next step, Long and Kuhl turned to machine-learning algorithms to learn from the neural activity patterns and index the states of the brain, regardless of the instruction on which objects, old or new, to focus on. With this information, the researchers were able to detect brain states which predicted the expression of memories in the parietal cortex.

“This validates the idea that there is a segregation of memory-related information versus perception,” Kuhl said. “This experience is common. When you are thinking about something else while driving to work, your eyes are open, you drive a few blocks, but you don’t remember those blocks because your consciousness was on the past.”

In the Journal of Neuroscience, Kuhl’s doctoral student Yufei Zhao found that MRI-observed activity in the parietal cortex of 29 subjects reflected an exaggeration effect that reduces interference when encoding subtle differences between newly seen objects. That divergent brain activity patterns strengthens memory, the study concluded.

Study participants viewed images of objects paired with photos of faces. Some of the objects were identical but with slight color differences. After practice and testing the tasks, the subjects’ parietal cortex activity was observed with MRI as they were shown a face and asked to adjust a color wheel to match which object corresponded to the face.

The research team uncovered a consistent systematic error. The color differences displayed on the wheel constantly were always 24 degrees apart, but subjects systematically remembered the differences as being larger.

“The further apart that they remembered the colors as being, the more likely they were to remember the faces associated with the objects,” Kuhl said. “What we show is that when we look at activity patterns in this part of the brain, the less similar those activity patterns are, the greater the exaggeration in memory. This exaggerated separation helps to avoid overlapping memory recall and avoid confusion.”

Kuhl, who is in his third year of research under a five-year National Science Foundation Career Award, sees himself as a middleman in neuroscience.

“I don’t work directly with patients with disease, but I am really interested in how the brain supports our normal functions with memory,” he said. “It’s incredibly complex. We’ve learned a lot, but there are a lot of things we don’t know. I think this problem of how we can avoid interference from memories has been around a long time.”

The two studies were funded by the National Institutes of Health, National Science Foundation and Lewis Family Endowment, which supports the UO’s Robert and Beverly Lewis Center for Neuroimaging.

—By Jim Barlow, University Communications